Overview
Cluster administrators can use the kubectl command line interface (CLI) tool to debug pods, services, and other issues in Kubernetes clusters. This article describes some basic kubectl commands to debug pods in your cluster.
Prerequisites
- One or more pods are running in a cluster.
- You have sufficient permissions to access the cluster and execute commands using the
kubectlCLI.
Common commands for debugging pods
As a best practice, debug pods using the following sequence of operations:
- Display a list of pods and their status.
- Display the logs for specific or all containers in a pod.
- Execute commands in a specific container.
- Debug pods using their clones for interactive debugging.
This section describes the relevant commands that support the suggested sequence.
kubectl get pods
To get basic information about the list of available pods and their status, use the following command:
Syntax
$ kubectl get pods
You can also list pods for a specific namespace by using the --namespace option:
$ kubectl get pods --namespace=NAMESPACE
Example:
$ kubectl get pods --namespace=default
NAME READY STATUS RESTARTS AGE
hello-node-7b87cd5f68-2wp4m 1/1 Running 0 21m
Alternatively, you can access the pods from a browser-based Kubernetes dashboard:
Click a pod to display its details.
kubectl logs
To display a snapshot of the logs of containers running inside a pod, use the following command:
Syntax
$ kubectl logs POD --all-containers=true
To display a snapshot of logs for a specific container running in a pod, use the following command:
$ kubectl logs -p -c CONTAINER POD
Note: If a pod has a single container, the name of the container is accepted as the default argument.
Example
$ kubectl logs hello-node-7b87cd5f68-2wp4m
I0715 06:51:04.198447 1 log.go:195] Started HTTP server on port 8080
I0715 06:51:04.198572 1 log.go:195] Started UDP server on port 8081
kubectl exec
To execute a shell command in a specific container, use the following command:
Syntax
$ kubectl exec POD -c CONTAINER -- SHELL-COMMAND
By default, the command is executed in the first container of the pod if you do not provide the name of a specific container.
Example
Execute the date command in a container of the hello-node-7b87cd5f68-2wp4m pod:
$ kubectl exec hello-node-7b87cd5f68-2wp4m -- date
Sat Jul 15 08:05:14 UTC 2023
kubectl debug
To debug cluster resources using interactive debugging containers, use the following command:
Syntax
$ kubectl debug POD -it --image=IMAGE
Example
$ kubectl debug hello-node-7b87cd5f68-2wp4m -it --image=busybox
Defaulting debug container name to debugger-55rjd.
If you don't see a command prompt, try pressing enter.
/ #
/ #
/ # pwd
/
/ # ls -a
. .dockerenv dev home lib64 root tmp var
.. bin etc lib proc sys usr
/ # ls -alrt
total 4
lrwxrwxrwx 1 root root 3 Jun 24 00:04 lib64 -> lib
drwxr-xr-x 1 root root 270 Jun 24 00:04 lib
drwxr-xr-x 1 root root 4758 Jun 24 00:04 bin
drwxr-xr-x 1 root root 16 Jun 24 00:04 var
drwxr-xr-x 1 root root 14 Jun 24 00:04 usr
drwxrwxrwt 1 root root 0 Jun 24 00:04 tmp
drwxr-xr-x 1 nobody nobody 0 Jun 24 00:04 home
dr-xr-xr-x 13 root root 0 Jul 15 06:50 sys
dr-xr-xr-x 565 root root 0 Jul 15 08:18 proc
drwxr-xr-x 1 root root 56 Jul 15 08:18 etc
drwxr-xr-x 5 root root 380 Jul 15 08:18 dev
-rwxr-xr-x 1 root root 0 Jul 15 08:18 .dockerenv
drwxr-xr-x 1 root root 8 Jul 15 08:18 ..
drwxr-xr-x 1 root root 8 Jul 15 08:18 .
drwx------ 1 root root 24 Jul 15 08:19 root
/ # exit
Session ended, the ephemeral container will not be restarted but may be reattached using 'kubectl attach hello-node-7b87cd5f68-2wp4m -c debugger-55rjd -i -t' if it is still running
Additional resources
For detailed information on debugging pods in a Kubernetes cluster, refer to the following resources: